Young songbirds learn to produce complex vocalizations (song) from their adult tutor during a critical period of development, just as human infants learn to speak from their parents. Since song learning critically depends on a discrete and specialized basal ganglia-thalamo-cortical circuit called the anterior forebrain pathway (AFP), songbirds offer a great and rare opportunity to study the mechanisms underlying basal ganglia-dependent learning of complex motor skills in early life. One of the major obstacles in studying song learning mechanisms is complexity of the behavior. Juvenile birds gradually develop their song from unstructured and highly variable vocalizations akin to human babbling, by changing many acoustic and temporal features all together. This highly complex learning makes it difficult to examine precise functions of neural structures and activity in behavior. Recent studies have overcome this difficulty by using much simpler learning of adult song, which is experimentally induced using an operant conditioning paradigm with aversive stimuli and is targeted to a specific feature of song (e.g., pitch of a particular portion of song). Althouh these studies have revealed important functions of the AFP in adult song learning, it is still unknown whether the same functions are employed for the much more complex, self-driven learning of juvenile song. Indeed, several recent findings, including my own preliminary data, even raise the possibility that juvenile song learning is substantially different from that of aduls in its mechanisms. Thus, examining whether the findings in adult song learning are generally applicable to juvenile song learning is critically important for understanding the mechanisms of song learning, and an answer either way will be a significant advance in the birdsong research field. This proposal specifically addresses this issue by directly comparing several aspects of AFP's function in juvenile song learning with those in adult song learning reported previously. To do so, I will use new behavioral manipulations of juvenile learning that allow us to precisely analyze the same features of juvenile song as those analyzed in adult song learning (both pitch and sequence), and thus enable us to examine the AFP's function in the same way as that in the adult learning studies. This study will not only test whether adults provide a faithful model fr studying mechanisms of juvenile song learning, but will also show that detailed analyses of song learning mechanisms can be done even in juveniles, using targeted manipulations of learning - this provides a new and powerful approach to the study of vocal learning mechanisms in songbirds. Moreover, this study has the potential to reveal differences in neural mechanisms between self-driven, age-dependent development of complex motor skills in young animals such as speech acquisition in human infants, and simple, training-induced changes of well-learned adult behavior such as fine adjustment of speech production in adults.